Find research datasets worth reusing
Search datasets from major research repositories and use ShareScore to quickly assess how well each record supports discovery, access, and reuse.
171
datasets available to search
ShareScore release 0.9.0
Dataset results
171 results for “Model inference”
Data of Bayesian inference of non-linear multiscale model parameters accelerated by a Deep Neural Network
<pre>Data from title = "Bayesian inference of non-linear multiscale model parameters accelerated by a Deep Neural Network", journal = "Computer Methods in Applied Mechanics and Engineering", pages = "112693", year = "2020", issn = "0045-7825", doi = "https://doi.org/10.1016/j.cma.2019.112693", author = "Wu, Ling and Zulueta, Kepa and Major, Zoltan and Arriaga, Aitor and Noels, Ludovic" </pre>
MLPerf Inference BERT sparse model on SQuAD v1.1 dataset
<p>BERT Large.</p> <p>{<br> "hidden-size": 1024,<br> "num-attention-heads": 16,<br> "num-layers": 24,<br> "max-seq-length": 384<br> }</p> <p>Fine tuned model for sparsity.</p>
Data for "Bayesian inference for biophysical neuron models enables stimulus optimization for retinal neuroprosthetics"
<p>Experimental and precomputed data for the paper "Bayesian inference for biophysical neuron models enables stimulus optimization for retinal neuroprosthetics" by Oesterle et al. 2020 (DOI: <a href="https://doi.org/10.7554/eLife.54997">10.7554/eLife.54997</a>).</p> <p>The cone bipolar cell data has been described and published in the paper "Inhibition decorrelates visual feature representations in the inner retina" by Franke et al. 2017 (DOI: <a href="https://doi.org/10.1038/nature21394">10.1038/nature21394</a>). </p> <p>This data is both a supplement to the Oesterle et al. paper and the code for this paper.</p> <p>The code is available in this <a href="http://github.com/berenslab/CBC_inference">GitHub repository</a>.</p> <p>We recommend downloading the GitHub repository and to follow the instructions there.</p>
Data from: Validating dispersal distances inferred from autoregressive occupancy models with genetic parentage assignments
1.Dispersal distances are commonly inferred from occupancy data but have rarely been validated. Estimating dispersal from occupancy data is further complicated by imperfect detection and the presence of unsurveyed patches. 2.We compared dispersal distances inferred from seven years of occupancy data for 212 wetlands in a metapopulation of the secretive and threatened California black rail (Laterallus jamaicensis coturniculus) to distances between parent-offspring dyads identified with 16 microsatellites. 3.We used a novel autoregressive multi-season occupancy model that accounted for both unsurveyed patches and imperfect detection to quantify patch isolation using buffer radius (BRM) and incidence function (IFM) connectivity measures at 15 scales (1–10, 15, 20, 25, and 30 km). Connectivity measures were then fit as colonization covariates in occupancy models to estimate a model-averaged dispersal distance. 4.As predicted, colonization was more strongly related to connectivity at small spatial scales (< 10 km). AIC weights were greatest at 7 km for BRM and at 4 km for IFM. 5.Model-averaged dispersal distances (BRM = 7.46 km; IFM = 5.48 km) showed good agreement with the mean (± SE) dispersal distance from 23 parent-offspring dyads (5.58 ± 1.92 km), indicating reasonably accurate mean dispersal distances can be inferred from occupancy data when isolation strongly affects colonization.
Bayesian inference of ancestral host-parasite interactions under a phylogenetic model of host repertoire evolution
<p>Intimate ecological interactions, such as those between parasites and their hosts, may persist over long time spans, coupling the evolutionary histories of the lineages involved. Most methods that reconstruct the coevolutionary history of such interactions make the simplifying assumption that parasites have a single host. Many methods also focus on congruence between host and parasite phylogenies, using cospeciation as the null model. However, there is an increasing body of evidence suggesting that the host ranges of parasites are more complex: that host ranges often include more than one host and evolve via gains and losses of hosts rather than through cospeciation alone. Here, we develop a Bayesian approach for inferring coevolutionary history based on a model accommodating these complexities. Specifically, a parasite is assumed to have a host repertoire, which includes both potential hosts and one or more actual hosts. Over time, potential hosts can be added or lost, and potential hosts can develop into actual hosts or vice versa. Thus, host colonization is modeled as a two-step process that may potentially be influenced by host relatedness. We first explore the statistical behavior of our model by simulating evolution of host-parasite interactions under a range of parameter values. We then use our approach, implemented in the program RevBayes, to infer the coevolutionary history between 34 Nymphalini butterfly species and 25 angiosperm families. Our analysis suggests that host relatedness among angiosperm families influences how easily Nymphalini lineages gain new hosts.</p>
Data for simultaneous inference of sea ice state and surface emissivity model using machine learning and data assimilation
<h2>Overview</h2> <p>This dataset supports the draft manuscript "Simultaneous inference of sea ice state and surface emissivity model using machine learning and data assimilation" which describes a way to infer the daily maps of the sea ice concentration and empirical properties of the sea ice (relating to its snow cover and its physical properties, such as air inclusions) along with the creation of a new empirical model for the sea ice surface emissivity. This is done using knowledge of the atmosphere state, skin temperature and ocean water emissivity from the European Centre for Medium-range Weather Forecasts (ECMWF) weather forecasting model and the observed radiances at microwave frequencies from the Advanced Microwave Scanning Radiometer 2 (AMSR2). The inverse modelling and state estimation is achieved by combining empirical machine learning elements in a Bayesian-inspired network along with a number of physical components. The work also introduces the idea of an "empirical state", in this case describing the aspects of the sea ice physical state which affect the observations, and which is defined by the inputs to the new empirical model component (in machine learning terms, it is defined by the latent input state of a neural network). This dataset includes the data used in training the model and inferring the sea ice parameters, as well as the outputs from that training process. The software used to perform the training is in Python and uses the Keras and Tensorflow software. See the draft manuscript for full details of this data.</p> <p>The code used in the draft manuscript is archived at <a href="https://doi.org/10.5281/zenodo.10013542">https://doi.org/10.5281/zenodo.10013542</a></p> <p>The data used in the draft manuscript is archived at <a href="https://doi.org/10.5281/zenodo.10033377">https://doi.org/10.5281/zenodo.10033377</a></p> <h2>Training data </h2> <h3>Observation space training and ancillary data</h3> <p>Training is done at the location of AMSR2 superobservations (superobs) over ocean with less than 1% land contamination and polewards of 45 degrees latitude, between 1st July 2020 and 30th June 2021. There are 64,184,021 superobs used. A superob is the average of all raw JAXA level 1B observations from one orbit falling into a grid box on an approximately constant area (reduced Gaussian) grid at approximately 40 km by 40 km resolution (noting that polar regions can thus have up to around 7 superobs per day). The superobs have been computed using the field of view central locations for each channel as derived from the JAXA level 1B data. A subset of 10 of the AMSR2 channels is used, from 10 GHz, V polarised, to 89 GHz, H polarised.</p> <p>At each superob location, the relevant fields from the ECMWF 12 hour 'background' forecast are interpolated to the observation time and location. The atmosphere is represented indirectly by the relevant radiative transfer terms from a scattering radiative transfer model. The sea ice concentration from the ECMWF OCEAN5 analysis is included as a validation reference but is not used in the training itself, except to provide a monthly mean first guess to speed up the training. Each field is provided in a separate netCDF file:</p> <ul> <li>field_v2_JULIAN_DAY.nc - superob time in days since 12 UTC on Nov 24th 4714 BC on the proleptic Gregorian calendar</li> <li>field_v2_LAT.nc - superob central latitude in degrees</li> <li>field_v2_LON.nc - superob central longitude in degrees</li> <li>field_v2_IGRID.nc - corresponding grid number on the map grid used in this work (see below)</li> <li>field_v2_OBSVALUE.nc - observed superob brightness temperature at each of 10 AMSR2 channels.</li> <li>field_v2_TSFC.nc - skin temperature computed by the ECMWF forecast model</li> <li>field_v2_WINDSPEED10M.nc - 10m wind speed computed by the ECMWF forecast model</li> <li>field_v2_EMIS_WATER.nc - Ocean water surface emissivity at 10 AMSR2 channels, simulated from the ECMWF forecast fields using the FASTEM-6 model</li> <li>field_v2_CLOUD_FRACTION.nc - Effective cloud fraction used in the atmospheric radiative transfer model at each of 10 AMSR2 channels</li> <li>field_v2_TAUSFC_CLD.nc - Surface to space transmittance in the cloudy column at each of 10 AMSR2 channels</li> <li>field_v2_TUP_CLD.nc - Upwelling brightness temperature from the atmosphere in the cloudy column at each of 10 AMSR2 channels</li> <li>field_v2_TDOWN_CLD.nc - Downwelling brightness temperature from the atmosphere in the cloudy column at each of 10 AMSR2 channels</li> <li>field_v2_TAUSFC.nc - Equivalently for the clear column</li> <li>field_v2_TUP.nc - Equivalently for the clear column</li> <li>field_v2_TDOWN.nc - Equivalently for the clear column</li> <li>field_v2_SEAICE.nc - Sea ice concentration from the ECMWF OCEAN5 analysis, for validation only (not used in training)</li> </ul> <h3>Grid space data: initial data for training; validation sea ice data</h3> <p>A number of properties are provided to the hybrid physical-empirical model that is being trained, on a special map grid defined in this project, including all 62,499 of the reduced Gaussian 40km grid points that have at least one superob at some point during the year of training data. These are:</p> <ul> <li>ifs_seaice_initials_year.nc - sea ice concentration from OCEAN5, monthly averaged on the grid, and then provided on all days of the relevant month as initial conditions (technically, first guess) for the training. This includes an additional day before the beginning of the training, used for time-lagging (see draft paper).</li> <li>ifs_tsfc_year_dailyx.nc - skin temperature from ECMWF forecast fields at observation locations, averaged onto the daily grid, to help provide constraints on the likelihood of sea ice as part of a sea ice loss function.</li> </ul> <p>For diagnostic and validation purposes, the ECMWF OCEAN5 analysis is also provided on the grid:</p> <ul> <li>ifs_seaice_year.nc - sea ice concentration from OCEAN5 at observation locations, averaged onto the daily grid</li> </ul> <p>All these fields are provided on the following dimensions:</p> <ul> <li>LON - the longitude of the grid point in degrees</li> <li>DAY - the day through the training year (0-364, 1st July 2020 to 30th June 2021) or through the training year extended forward by one day (30th June 2020) for the sea ice (0-365). In practice the days are offset by 3 hours from the UTC day to match the ECMWF data assimilation windows, which start at 21 UTC the day before.</li> </ul> <p>The latitude is also provided</p> <ul> <li>LAT - the latitude of the grid point in degrees</li> </ul> <p>Note that the observation location IGRID is on the custom grid of the ML model that is defined implicitly in these gridded files. The LON and LAT vectors in these files are the longitude and latitude points of the grid and are of 62499 in length. The IGRID number for an observation is the index into these arrays from 0-62498.</p> <h2>Outputs from training</h2> <p>The following files are the output and diagnostics from the year-long training. The python code and the draft paper are the primary documentation for these:</p> <ul> <li>models_year.nc - settings of the model are recorded here, along with the trained values of the smaller empirical components/layers within the hybrid model. For example, the layer weights of the wind speed bias correction, the observation space bias correction, and the empirical surface emissivity model are recorded here. The values of the loss function at each epoch are also recorded here.</li> <li>properties_year.nc - trained values of each of 3 empirical properties of sea ice on the map grid (3 properties by 62499 locations by 365 days from 1st July 2020)</li> <li>seaice_year.nc - inferred values of sea ice fraction on the map grid (62499 locations by 365 days from 1st July 2020, discarding the additional day at the start)</li> <li>tbsim_year.nc - simulated AMSR2 brightness temperatures from the trained network</li> <li>tbsim_initial_year.nc - simulated AMSR2 brightness temperatures using the untrained network</li> </ul> <p>The longitude and latitude of the map grid is found in any of the initial data files described in the previous section. The days are 0-364 corresponding to 1st July 2020 to 30th June 2021.</p> <h3>Sea ice surface emissivity at grid locations</h3> <p>A packaged version of the sea ice surface emissivity is provided at grid locations, alongside the surface emissivity model, the sea ice concentration and the four inputs to the model, i.e. the normalised skin temperature and the three empirical variables:</p> <ul> <li>emissivity_grid_year.nc</li> </ul> <p>Note that in the training, the surface emissivity is computed at observation locations and has not been stored due to memory limitations. For easier comparison to other datasets, the surface emissivity has been recomputed on grid locations in this package, using the year-long trained emissivity model and its trained inputs. The sea ice surface emissivity is only physically meaningful for sea ice concentrations above around 0.25. Also be aware of the "hole at the pole" which is the small region of the Arctic ocean that is sometimes not covered by an AMSR2 overpass, and which is found from 88 degrees N. On days where the hole or part of the hole exists, the sea ice emissivity on the grid is not valid at these locations. These locations can be identified by having all values of the empirical properties zero (because the empirical properties were never constrained by any observations on that day, and remain at their initial values before training).</p> <h2>Sensitivity tests</h2> <p>Extensive sensitivity tests were carried out, as described in the appendices of the draft paper and as documented in the Python code, using the month of August 2020 as an example. These required equivalent month-long training and initial data similar to those described above, but all observation space fields are contained within the same file in this case. Output files follow similar principles to those described above. The full package is provided as a tar file:</p> <ul> <li>sensitivity.tar</li> </ul> <p>This contains the training and initial files:</p> <ul> <li>amsr2_v2_202008.nc</li> <li>ifs_tsfc_dailyx_202008.nc</li> <li>ifs_seaice_202008.nc</li> </ul> <p>as well as directories containing the trained model outputs and diagnostics at each of the sensitivity tests, using the same formats as described for the yearly training, with these names:</p> <ul> <li>nprop - number of empirical properties</li> <li>epoch - number of epochs</li> <li>deep - configuration of the empirical sea ice emissivity model, including multiple layers of nonlinear dense neural network</li> <li>bseaice - background error for the sea ice physical bounds background error (loss) term</li> <li>bemis - background error for the sea ice emissivity background error (loss) term</li> <li>bbias - background error for the bias correction background error (loss) term</li> <li>batchsize - batch size used in training</li> <li>bbatchsize - extended epochs testing of batch size used in training</li> </ul> <h2>Licensing</h2> <p>This data product is published under a Creative Commons Attribution 4.0 International (CC BY 4.0). To view a copy of this licence, visit <a href="https://creativecommons.org/licenses/by/4.0/">https://creativecommons.org/licenses/by/4.0/</a></p> <p>You are free to:</p> <ul> <li>Share — copy and redistribute the material in any medium or format</li> <li>Adapt — remix, transform, and build upon the material for any purpose, even commercially.</li> </ul> <p>Under the following terms:</p> <ul> <li>You must give appropriate credit (attribution) to ECMWF as outlined below, provide a link to the licence, and indicate if changes were made.</li> <li>No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the licence permits.</li> </ul> <p>The following wording shall be attached to the use of this ECMWF data product: </p> <ol> <li>Copyright statement: Copyright "© 2023 European Centre for Medium-Range Weather Forecasts (ECMWF)".</li> <li>Source <a href="http://www.ecmwf.int/">www.ecmwf.int </a>and <a href="https://doi.org/10.5281/zenodo.10009497">https://doi.org/10.5281/zenodo.10009497</a></li> <li>Licence Statement: This data is published under a Creative Commons Attribution 4.0 International (CC BY 4.0). <a href="https://creativecommons.org/licenses/by/4.0/">https://creativecommons.org/licenses/by/4.0/</a></li> <li>Disclaimer: ECMWF does not accept any liability whatsoever for any error or omission in the data, their availability, or for any loss or damage arising from their use.</li> <li>Where applicable, an indication if the material has been modified and an indication of previous modifications.</li> <li>DOI: 10.5281/zenodo.10009498</li> </ol> <p>Original data for this value-added product was provided by Japan Aerospace Exploitation Agency (JAXA). Specifically, this dataset builds on the Advanced Microwave Scanning Radiometer 2 (AMSR2) level 1B data available from the JAXA G-Portal, https://gportal.jaxa.jp/gpr/, which has the following attribution and licensing:</p> <ol> <li>Give credit for the original data to JAXA, i.e. "Original data for this value added data product was provided by Japan Aerospace Exploration Agency"</li> <li>DOI for original JAXA data is L1B-Brightness temperature (TB) GCOM-W/AMSR2 L1B Brightness Temperature: <a href="https://doi.org/10.57746/EO.01gs73ans548qghaknzdjyxd2h">https://doi.org/10.57746/EO.01gs73ans548qghaknzdjyxd2h</a></li> <li>Original terms of data service from JAXA, with highlighted extracts: <ul> <li><a href="https://gportal.jaxa.jp/gpr/index/eula?lang=en">https://gportal.jaxa.jp/gpr/index/eula</a> <ol> <li>The user is entitled to use G-Portal data free of charge without any restrictions (including commercial use) except for the condition about acknowledgement of data credit as stipulated in Article 7.(2). (see above)</li> <li>JAXA is collecting results (papers, theses, reports, etc.) using G-Portal data. If you have any results using G-Portal data, please mail/e-mail a copy of the result to G-Portal Support Desk (Contact Information written at the end of the Terms of Use). We appreciate your cooperation very much.</li> </ol> </li> </ul> </li> </ol>
Supplementary files from: Improving inference and avoiding over-interpretation of hidden-state diversification models: Specialized plant breeding has no effect on diversification in frogs
<p>The hidden-state speciation and extinction (HiSSE) model helps avoid spurious results when testing whether a character affects diversification rates. However, care must be taken to optimally analyze models and interpret results. Recently, Tonini et al. (2020; TEA hereafter) studied anuran (frog and toad) diversification with HiSSE methods. They concluded that their focal state, breeding in phytotelmata, increases net diversification rates. Yet this conclusion is counterintuitive, because the state that purportedly increases net diversification rates is 14 times rarer among species than the alternative. Herein I revisit TEA's analyses and demonstrate problems with inferring model likelihoods, conducting post-hoc tests, and interpreting results. I also re-evaluate their top models and find that diverse strategies are necessary to reach the parameter values that maximize each model's likelihood. In contrast to TEA, I find no support for an effect of phytotelm breeding on net diversification rates in Neotropical anurans. In particular, even though the most highly supported models include the focal character, averaging parameter estimates over hidden states shows that the focal character does not influence diversification rates. Finally, I suggest ways to better analyze and interpret complex diversification models – both state-dependent and beyond – for future studies in other organisms.</p>
Impact of model assumptions on the inference of the evolution of ectomycorrhizal symbiosis in fungi
<p>Ectomycorrhiza (ECM) is a symbiotic relation between plant and fungi that is essential for nutrient uptake of many stand forming trees. There are two conflicting views about the evolution of ECM in fungi suggesting (1) relatively few transitions to ECM followed by reversals to non-ECM, or (2) many independent origins of ECM and no reversals. In this study, we compare these, and other, hypotheses and test the impact of different models on inference. We assembled a dataset of five marker gene sequences (nuc58, nucLSU, nucSSU, rpb1, and rpb2) and 2,174 fungal taxa covering the three subphyla: Agaricomycotina, Mucoromycotina and Pezizomycotina. The fit of different models, including models with variable rates in clades or through time, to the pattern of ECM fungal taxa was tested in a Bayesian framework, and using AIC and simulations. We find that models implementing variable rates are a better fit than models without rate shift, and that the conclusion about the relative rate between ECM and non-ECM depend largely on whether rate shifts are allowed or not. We conclude that standard constant-rate ancestral state reconstruction models are not adequate for the analysis of the evolution of ECM fungi, and may give contradictory results to more extensive analyses. </p>
Data from: Sharing detection heterogeneity information among species in community models of occupancy and abundance can strengthen inference
<p>1. The estimation of abundance and distribution and factors governing patterns in these parameters is central to the field of ecology. The continued development of hierarchical models that best utilize available information to inform these processes is a key goal of quantitative ecologists. However, much remains to be learned about simultaneously modeling true abundance, presence, and trajectories of ecological communities.</p> <p>2. Simultaneous modeling of the population dynamics of multiple species provides an interesting mechanism to examine patterns in community processes and, as we emphasize herein, to improve species-specific estimates by leveraging detection information among species. Here we demonstrate a simple but effective approach to share information about observation parameters among species in hierarchical community abundance and occupancy models, where we use shared random effects among species to account for spatiotemporal heterogeneity in detection probability.</p> <p>3. We demonstrate the efficacy of our modeling approach using simulated abundance data, where we recover well our simulated parameters using N-mixture models. Our approach substantially increases precision in estimates of abundance compared to models that do not share detection information among species. We then expand this model, and apply it to repeated detection/non-detection data collected on six species of tits (Paridae) breeding at 119 1 km<sup>2</sup> sampling sites across a <em>P. montanus</em> hybrid zone in northern Switzerland (2004-2020). We find strong impacts of forest cover and elevation on population persistence and colonisation in all species. We also demonstrate evidence for interspecific competition on population persistence and colonization probabilities, where the presence of marsh tits reduces population persistence and colonisation probability of sympatric willow tits, potentially decreasing gene flow among willow tit subspecies.</p> <p>4. While conceptually simple, our results have important implications for the future modeling of population abundance, colonization, persistence, and trajectories in community frameworks. We suggest potential extensions of our modeling in this paper, and discuss how leveraging data from multiple species can improve model performance and sharpen ecological inference.</p>
Integrated animal movement and spatial capture-recapture models: simulation, implementation, and inference
<p>Over the last decade, spatial capture-recapture (SCR) models have become widespread for estimating demographic parameters in ecological studies. However, the underlying assumptions about animal movement and space use are often not realistic. This is a missed opportunity because ecological questions related to animal space use, habitat selection, and behavior cannot be addressed with most SCR models, despite the fact that the data collected in SCR studies -- individual animals observed at specific locations and times -- can provide a rich source of information about how these processes relate to demographic rates. We developed SCR models that integrate complex movement processes that are typically inferred from telemetry data, including a simple random walk, correlated random walk (i.e., short-term directional persistence), and habitat-driven Langevin diffusion. We demonstrated how to formulate, simulate from, and fit these models with standard SCR data using Bayesian analysis methods. We evaluated their performance through a simulation study, where we varied the detection, movement, and resource selection parameters. We also examined different numbers of sampling occasions and assessed performance gains when including auxiliary location data collected from telemetered individuals. Across all scenarios, the integrated SCR movement models performed well in terms of abundance, detection, and movement parameter estimation. We found little difference in bias for the simple random walk model when reducing the number of sampling occasions from T=25 to T=15. We found some bias in movement parameter estimates under several of the correlated random walk scenarios, but incorporating auxiliary location data improved parameter estimates and significantly improved mixing during model fitting. The Langevin movement model was able to recover resource selection parameters from standard SCR data, which is appealing because it explicitly links the individual-level movement process with habitat selection and population density. We focused on closed population models, but movement models developed here could be extended to open SCR models. The movement process models could also be extended to accommodate additional "building blocks'' of random walks, such as central tendency (e.g., territoriality) or multiple movement behavior states, thereby providing a flexible and coherent framework for linking animal movement behavior to population dynamics, density, and distribution.</p>
The planted XY model: thermodynamics and inference
<p>Data and codes for figures in manuscript "The planted XY model: thermodynamics and inference"</p>
SonicParanoid2: fast, accurate, and comprehensive orthology inference with machine learning and language models
<p>This repository contains the documentation, test datasets and scripts used in the following study:</p> <p>"SonicParanoid2: fast, accurate, and comprehensive orthology inference with machine learning and language models"<br><br>- `sonic-manuscript-master.zip` contains all the scripts to reproduce the study, including those for generating the figures and tables included in the manuscript.</p> <p>- <a href="../api/records/11361985/draft/files/sonicparanoid2.wiki.tar.xz/content" target="_blank" rel="noopener noreferrer">sonicparanoid2.wiki.tar.xz</a> contains a snapshot fo the wiki for SonicParanoid2 as of May 30, 2024</p>
Data from: Integrating niche and occupancy models to infer the distribution of an endemic fossorial snake (Atractus lasallei)
<p>Understanding species distribution and habitat preferences is crucial for effective conservation strategies. However, the lack of information about population responses to environmental change at different scales hinders effective conservation measures. In this study, we estimate the potential and realized distribution of <em>Atractus lasallei</em>, a semi-fossorial snake endemic to the northwestern region of Colombia. We modelled the potential distribution of <em>A. lasallei</em> based on ecological niche theory (using maxent), and habitat use was characterized while accounting for imperfect detection using a single-season occupancy model. Our results suggest that <em>A. lasallei</em> selects areas characterized by slopes below 10°, with high average annual precipitation (>2500mm/year) and herbaceous and shrubby vegetation. Its potential distribution encompasses the northern Central Cordillera and two smaller centers along the Western Cordillera, but its habitat is heavily fragmented within this potential distribution. When the two models are combined, the species' realized distribution sums up to 935 km<sup>2</sup>, highlighting its vulnerability. We recommend approaches that focus on variability at different spatio-temporal scales to better comprehend the variables that affect species' ranges and identify threats to vulnerable species. Prompt actions are needed to protect herbaceous and shrub vegetation in this region, highly demanded for agriculture and cattle grazing.</p>
Fig. 2 in Phylogeography and potential glacial refugia of terrestrial gastropod Faustina faustina (Rossmässler, 1835) (Gastropoda: Eupulmonata: Helicidae) inferred from molecular data and species distribution models
Fig. 2 Haplotype distribution for nuclear markers: ITS-2 (left) and 28S rRNA (right)
Delta GUI change detection using inferred models replication package
<p><strong>TESTAR </strong>is a scriptless automated open-source tool developed by the Universitat Politècnica de València and the Open University of the Netherlands.</p> <p><strong>TESTAR Change detection .NET</strong> is an open-source tool that simultaneously transits and compares state models inferred by a scriptless testing tool, enabling the detection and highlighting of GUI changes to detect the widgets or functionalities that have been added, removed, or modified. This tool is also developed by the Universitat Politècnica de València and the Open University of the Netherlands.</p> <p>This replication package contains:</p> <ul> <li>The Excel file that was used to perform and store the systematic mapping of the literature.</li> <li>The TESTAR Change detection .NET version that was used to compare the state models inferred from the OBS, Calibre, and MyExpenses applications. </li> <li>The state models that were inferred from the OBS, Calibre, and MyExpenses applications. These are stored in the OrientDB graph database. </li> <li>Three documents (OBS, Calibre, MyExpenses) that detail with images the GUI changes results detected using the TESTAR Change detection .NET tool.</li> <li>A video demo that shows how to use the TESTAR Change detection .NET tool with the inferred models from the Calibre web system.</li> </ul>
How to select predictive models for decision making or causal inference? Experiments data
<p>This is the full result data for the experiments of the paper : Doutreligne, M., & Varoquaux, G. (2023). How to select predictive models for decision making or causal inference?, https://hal.science/hal-03946902. <br><br>The code repository is : https://github.com/soda-inria/caussim/tree/main</p> <p>The files in this dataset are the one for the most computationnally costly experiments. There is one folder for each of the four datasets used in the paper. Then, one folder for each of the experimental setup. The files required for the main figure (Fig.3) of the paper are the one labelled #fig3 in the following descriptions.</p> <p>Details on the files : </p> <p>.<br>├── acic_2016_save<br>│ ├── acic_2016__nuisance_non_linear__candidates_hist_gradient_boosting__dgp_1-77__rs_1-5<br>│ │ └── run_logs.csv: results for the experiment with non linear models for both the nuisances and the candidates<br>│ ├── acic_2016__nuisance_non_linear__candidates_ridge__dgp_1-77__rs_1-10<br>│ │ └── run_logs.csv: results for the experiment with non linear models for the nuisances and linear models for the candidates<br>│ └── acic_2016__stacked_regressor__dgp_1-77__seed_1-10<br>│ └── run_logs.csv: results for the experiment with stacked models (linear and non linear) for the nuisances and non linear models for the candidates #fig3<br>├── acic_2018_save<br>│ └── acic_2018__nuisance_non_linear__candidates_hist_gradient_boosting__first_uid_432<br>│ └── run_logs.csv results for the experiment with stacked models (linear and non linear) for the nuisances models and non linear models for the candidates #fig3<br>├── caussim_save<br>│ ├── caussim__linear_regressor__test_size_5000__n_datasets_1000<br>│ │ ├── run_logs.csv: results for the experiment with stacked models for the nuisances models and linear models for the candidates <br>│ │ └── simu.yaml: configuration file of the experiment<br>│ ├── caussim__nuisance_non_linear__candidates_ridge__overlap_01-247_join_nuisance_train_set<br>│ │ └── run_logs.csv: results for the experiment with non linear models for the nuisances and linear models for the candidates, joined sets for the nuisances and the candidates<br>│ ├── caussim__nuisance_non_linear__candidates_ridge__overlap_01-247_separated_nuisance_train_set<br>│ │ └── run_logs.csv: results for the experiment with non linear models for the nuisances and linear models for the candidates, separated sets for the nuisances and the candidates<br>│ └── caussim__stacked_regressor__test_size_5000__n_datasets_1000<br>│ ├── run_logs.csv: results for the experiment with stacked models (linear and non linear) for the nuisances and linear models for the candidates #fig3<br>│ └── simu.yaml: configuration file of the experiment<br>└── twins_save<br> └── twins__stacked_regressor__rs_1-10__overlap_0.1-3<br> └── run_logs.csv: results for the experiment with stacked models (linear and non linear) for the nuisances and non linear models for the candidates #fig3</p>
Sequential Bayesian Inference of Finite-strain Visco-elastic Visco-plastic model parameters of 22-month aged PA12 bulk material printed along different directions
<p>These are the data related to aged PA12 (22 months) following the methodology described in the following publication in which non-aged PA12 has been tested:</p> <p>title = "Bayesian inference of high-dimensional finite-strain visco-elastic-visco-plastic model parameters for additive manufactured polymers and neural network based material parameters generator.",<br>journal = "International Journal of Solids and Structures",<br>year = "2023",<br>volume = "283",<br>pages = "112470",<br>doi = "10.1016/j.ijsolstr.2023.112470",<br>author = "Wu, Ling and Anglade, Cyrielle and Cobian, Lucia and Monclus, Miguel and Segurado, Javier and Karayagiz, Fatma and Freitas, Ubiratan and Noels Ludovic"</p> <p>Contrarily to the non-aged material, since high-strain-rate tests are not available, only 5 Maxwell's branches are considered herein.</p> <h1>Description</h1> <p>BI code and results of the inference of a pressure-dependent visco-elastic visco-plastic model developed in [NGU16] with a umat implementation in <a href="https://gitlab.uliege.be/moammm/moammmPublic/code/-/tree/main/MaterialModels/FiniteStrain/Finite_VEVP">https://gitlab.uliege.be/moammm/moammmPublic/code/-/tree/main/MaterialModels/FiniteStrain/Finite_VEVP</a></p> <p>The sequential BI is described in [WU23] The experimental protocol is reported in [COB22,COB22b] but is herein applied on aged PA12</p> <p>To run the BI you need the open source code <a href="https://gitlab.onelab.info/cm3/cm3Libraries" target="_blank" rel="nofollow noreferrer noopener">https://gitlab.onelab.info/cm3/cm3Libraries</a></p> <p>If you use these data or model, we would be grateful if you could cite the related papers:</p> <ul> <li>[WU23] L. Wu, C. Anglade, L. Cobian, M. Monclus, J. Segurado, F. Karayagiz, U. Santos Freitas, L. Noels, Bayesian inference of high-dimensional finite-strain visco-elastic-visco-plastic model parameters for additive manufactured polymers and neural network based material parameters generator, International Journal of Solids and Structures (2023) 112470: <a href="https://doi.org/10.1016/j.ijsolstr.2023.112470" target="_blank" rel="nofollow noreferrer noopener">https://doi.org/10.1016/j.ijsolstr.2023.112470</a></li> <li>[COB22] L. Cobian, M. Rueda-Ruiz, J.P. Fernandez-Blazquez, V. Martinez, F. Galvez, F. Karayagiz, T. Lück, J. Segurado, M.A. Monclus, Micromechanical characterization of the material response in a PA12-SLS fabricated lattice structure and its correlation with bulk behaviour, Polymer Testing 110 (2022) 107556: <a href="https://doi.org/10.1016/j.polymertesting.2022.107556" target="_blank" rel="nofollow noreferrer noopener">https://doi.org/10.1016/j.polymertesting.2022.107556</a> (in Open access)</li> <li>[COB22b] Data of “. Cobian, M. Rueda-Ruiz, J.P. Fernandez-Blazquez, V. Martinez, F. Galvez, F. Karayagiz, T. Lück, J. Segurado, M.A. Monclus, Micromechanical characterization of the material response in a PA12-SLS fabricated lattice structure and its correlation with bulk behaviour, Polymer Testing 110 (2022) 107556” <a href="http://dx.doi.org/10.5281/zenodo.6136935" target="_blank" rel="nofollow noreferrer noopener">http://dx.doi.org/10.5281/zenodo.6136935</a> (in Open access)</li> <li>[NGU16] V. D. Nguyen, F. Lani, T. Pardoen, X. Morelle, L. Noels, A large strain hyperelastic viscoelastic-viscoplastic-damage constitutive model based on a multi-mechanism non-local damage continuum for amorphous glassy polymers. International Journal of Solids and Structures 96 (2016): 192-216; <a href="https://dx.doi.org/10.1016/j.ijsolstr.2016.06.008" target="_blank" rel="nofollow noreferrer noopener">https://dx.doi.org/10.1016/j.ijsolstr.2016.06.008</a>, Open access: <a href="https://orbi.uliege.be/handle/2268/197898" target="_blank" rel="nofollow noreferrer noopener">https://orbi.uliege.be/handle/2268/197898</a></li> </ul> <h1>Directories</h1> <p>All the codes and experimental results are in three directories:</p> <ol> <li>Experiment_PA12_AGED: experimental data of aged material, see the README.txt in each subdirectory for details</li> <li>BayesianVE: BI of the visco-elastic parameters<br>2.1. PlotExperimentalCurves: to vizualize the experimental curves and prepare the observations for the BI in the VE range<br>2.1.1. Loadcase_H.py and Loadcase_V.py read experimental results and create Load_ExpVE_H.dat and Load_ExpVE_V.dat, which keep the experimental observations and loading conditions to perform the BI.<br>2.1.2. PrintDir_H & PrintDir_V subdirectories with the functions called by Loadcase_H.py and Loadcase_V.py<br>2.1.3. Load_ExpVE_H.dat and Load_ExpVE_V.dat created files with the observations and loading conditions to perform the BI<br>2.2. VE_V and VE_H: BI for viscoelastic properties of "V" specimen (VE_V) and "H" specimen (VE_H)<br>2.2.1. BI_allpos_sequence.py runs the BI using Predict_VETest.py and creates the MCMC_VE_....dat<br>2.2.2. WarmStart = True is used to restart an inference<br>2.2.3. MCMC_VE_....dat in the VE_V and VE_H directories are the BI results<br>2.3. CheckBayRes: to visualize predictions of a BI sample and experimental curves 2.3.1. MCMCRes.py is used to check the numerical predictions of a BI parameter sample (read last sample by default, V or H direction can be selected at line 7)<br>2.3.2. ResKGEmu.py plots the evolution of elastic properties with time<br>2.3.3. uses as input VE_V/MCMC_VE_....dat or VE_H/MCMC_VE_....dat<br>2.3.4. uses local ViscoElasticTest.py, line.geo, line. msh as interface with <a href="https://gitlab.onelab.info/cm3/cm3Libraries" target="_blank" rel="nofollow noreferrer noopener">https://gitlab.onelab.info/cm3/cm3Libraries</a> code<br>2.3.5. uses local functions plotExp.py<br>2.4. ViscoElasticTest.py, line.geo, line.msh: interface with <a href="https://gitlab.onelab.info/cm3/cm3Libraries" target="_blank" rel="nofollow noreferrer noopener">https://gitlab.onelab.info/cm3/cm3Libraries</a> code used by VE_V and VE_H to call the VEVP model</li> <li>BayesianVEVP: BI of the visco-elastic and visco-plastic parameters<br>3.1. PlotExperimentalCurves: to vizualize the experimental curves and prepare the observations for the BI in the VE-VP ranges<br>3.1.1. Loadcase_H.py and Loadcase_V.py read experimental results and create Load_ExpVEVP_H.dat and Load_ExpVEVP_V.dat, which keep the experimental observations and loading conditions to perform BI at the viscoplastic stage.<br>3.1.2. PrintDir_H & PrintDir_V subdirectories with the functions called by Loadcase_H.py and Loadcase_V.py<br>3.1.3. Load_ExpVEVP_H.dat and Load_ExpVEVP_V.dat created files with the observations and loading conditions to perform the BI<br>3.2. VP_V and VP_H: BI for viscoelastic-viscoplastic properties of "V" specimen (VP_V) and "H" specimen (VP_H)<br>3.2.1. BI_allpos_sequence.py runs the BI using Predict_VETest.py and creates the MCMC_VP_....dat<br>3.2.2. WarmStart = True is used to restart an inference 3.2.3. It starts from the VE prosterior as prior, see point 2, and generates a MCMC_VP_?<em>1Step.dat (? being H or V)<br>3.3. CheckBayRes: visualize predictions of a BI sample and experimental curves<br>3.3.1. MCMCRes.py is used to check the numerical predictions with 3 BI parameter samples (inclusing MAP, V or H direction can be selected at line 12) using the samples of BayesianVEVP/VP</em>?/MCMC_VP_?<em>1Step.dat (? being H or V)<br>3.3.2. plot_hist.py is used to plot histograms of all the inferred parameters using the samples of BayesianVEVP/VP</em>?/MCMC_VP_?<em>1Step.dat (? being H or V)<br>3.3.3. Plot_Prop.py plots joints histograms of the inferred parameters using the samples of BayesianVEVP/VP</em>?/MCMC_VP_?_1Step.dat (? being H or V) 3.3.4. ResKGEmu.py plots the evolution of elastic properties with time<br>3.4. VEVPTest.py: interface with <a href="https://gitlab.onelab.info/cm3/cm3Libraries" target="_blank" rel="nofollow noreferrer noopener">https://gitlab.onelab.info/cm3/cm3Libraries</a> code used by VP_V2Step and VP_H2Step to call the VEVP model</li> </ol> <h1>Figures (reference to the number in [WU23] but for aged PA12)</h1> <ul> <li>Fig. 5 (Selected observations): From directory BayesianVE/PlotExperimentalCurves/PrintDir_? (? being H or V), run python3 plotExp_T.py or plotExp_C.py</li> <li>Fig. 7: BayesianVEVP/CheckBayRes/Plot_Prop.py with direct = "V" and then with direct = "H" and with Var = [0,1,14,18,22,23,24,25]</li> <li>Fig. 8 (Predictions of 3 inference realisations): BayesianVEVP/CheckBayRes/MCMCRes.py with direct = "V" (requires <a href="https://gitlab.onelab.info/cm3/cm3Libraries" target="_blank" rel="nofollow noreferrer noopener">https://gitlab.onelab.info/cm3/cm3Libraries</a> code)</li> <li>Fig. 9 (Predictions of 3 inference realisations): BayesianVEVP/CheckBayRes/MCMCRes.py with direct = "H" (requires <a href="https://gitlab.onelab.info/cm3/cm3Libraries" target="_blank" rel="nofollow noreferrer noopener">https://gitlab.onelab.info/cm3/cm3Libraries</a> code)</li> <li>Fig. 14A: From directory BayesianVE/PlotExperimentalCurves/PrintDir_? (? being H or V), run python3 plotExp_T.py or plotExp_C.py</li> <li>Fig. 15B: BayesianVEVP/CheckBayRes/Plot_Prop.py with direct = "V", Var = [2,3,8,9,10,11,12,13] and [14,15,16,17,18,19,20,21]</li> <li>Fig. 16B: BayesainVEVP/CheckBayRes/plot_hist.py with direct = "V"</li> <li>Fig. 17B: BayesianVEVP/CheckBayRes/plot_hist.py with direct = "V"</li> <li>Fig. 18B: BayesianVEVP/CheckBayRes/Plot_Prop.py with direct = "H", Var = [2,3,8,9,10,11,12,13] and [14,15,16,17,18,19,20,21]</li> <li>Fig. 19B: BayesianVEVP/CheckBayRes/plot_hist.py with direct = "H"</li> <li>Fig. 20B: BayesianVEVP/CheckBayRes/plot_hist.py with direct = "H"</li> </ul> <p> </p> <p>This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 862015.</p>
Geodetic model of the March 2021 Thessaly seismic sequence inferred from seismological and InSAR data
<p>A selection of Sentinel-1 (S1) wrapped and unwrapped measurements used in this study (from "a" to "u" files in tiff format as indicated in the word file attached). S1 data were processed by using our own internally developed InSAR processing chain.<br> <br> Earthquakes data locations.</p> <p><br> </p> <p> </p>
A greenhouse experiment partially supports inferences of ecogeographic isolation from niche models of Clarkia sister species
<p><b>Premise: </b>Ecogeographic isolation, or geographic isolation caused by ecological divergence, is thought to be of primary importance in speciation, yet is difficult to demonstrate and quantify. To determine whether distributions are limited by divergent adaptation or historical contingency, the gold standard is to reciprocally transplant taxa between their geographic ranges. Alternatively, ecogeographic isolation is inferred from species distribution models and niche divergence tests based on widely available environmental and occurrence data.</p> <p><b>Methods: </b>We test for ecogeographic isolation between two sister species of California annual wildflowers, <i>Clarkia concinna</i> and <i>C. breweri</i>, with a hybrid approach. We use niche models to predict water availability as the major axis of ecological divergence and then test that with a greenhouse experiment. Specifically, we manipulate water availability in field soils for two populations of each species and predict higher fitness in conditions representing home habitats to those representing the environment of each's sister species.</p> <p><b>Key Results: </b>Water availability and soil representing <i>C. concinna</i> generally increased both species' fitness. Thus, water and soil may indeed limit <i>C. concinna</i> from colonizing the range of C. breweri, but not vice versa. We suggest that the competitive environment and pollinator availability, which are not directly captured with either approach, may be key biotic factors correlated with climate that contribute to unexplained ecogeographic isolation for <i>C. breweri</i>.</p> <p><b>Conclusions:</b> Ours is a valuable approach to assessing ecogeographic isolation, in that it balances feasibility with model validation, and our results have implications for species distribution modeling efforts geared towards predicting climate change responses.</p>
Including a spatial predictive process in band recovery models improves inference for Lincoln estimates of animal abundance
<p>Abundance estimation is a critical component of conservation planning, particularly for exploited species where managers set regulations to restrict harvest based on current population size. An increasingly common approach for abundance estimation is through integrated population modeling (IPM), which uses multiple data sources in a joint likelihood to estimate abundance and additional demographic parameters. Lincoln estimators are one commonly used IPM component for harvested species, which combine information on the rate and the total number of individuals harvested within an integrated band-recovery framework to estimate abundance at large scales.</p> <p>A major assumption of the Lincoln estimator is that banding and recoveries are representative of the whole population, which may be violated if major sources of spatial heterogeneity in survival or harvest rates are not incorporated into the model. We developed an approach to account for spatial variation in harvest rates using a spatial predictive process, which we incorporated into a Lincoln estimator IPM.</p> <p>We simulated data under different configurations of sample sizes, harvest rates, and sources of spatial heterogeneity in harvest rate to assess potential model bias in parameter estimates. We then applied the model to data collected from a field study of wild turkeys (<em>Meleagris gallapavo</em>) to estimate local and statewide abundance in Maine, USA.</p> <p>We found that the band recovery model that incorporated a spatial predictive process consistently provided estimates of adult and juvenile abundance with low bias across a variety of spatial configurations of harvest rate and sampling intensities. When applied to data collected on wild turkeys, a model that did not incorporate spatial heterogeneity underestimated the harvest rate in some sub-regions. Consistent with simulation results, this led to over-estimation of both local and statewide abundance.</p> <p>Our work demonstrates that a spatial predictive process is a viable mechanism to account for spatial variation in harvest rates and limit bias in abundance estimates. This approach could be extended to large-scale band recovery datasets and has applicability for the estimation of population parameters in other ecological models as well.</p>
ScienceDex guides
Understand access before you commit
These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.
Allen Brain Atlas
Allen Brain Atlas is an Allen Institute collection of brain map atlases, datasets, APIs, and analysis tools covering mouse, human, and non-human primate brain resources.
Annotated Behaviour and Observability Dataset (ABODe)
ABODe is a University of Edinburgh DataShare dataset for behavior classification in group-housed mice using home-cage video, identities, bounding boxes, ground-plate positions, and annotator labels.
DANDI Archive for NWB datasets
DANDI is a BRAIN Initiative archive for publishing and sharing neurophysiology data, including electrophysiology, optophysiology, and behavioral data packaged as NWB and related standards.
International Brain Laboratory public data
The International Brain Laboratory public data releases expose standardized mouse decision-making experiments, including Neuropixels recordings, widefield calcium imaging, behavior, and session metadata accessed through the ONE API.
OpenNeuro
OpenNeuro is a free, open platform for sharing neuroimaging datasets, with public search, dataset pages, and download paths for web, S3, DataLad, and the OpenNeuro CLI.